# How to Predict Customer Churn? This is related with [[Machine Learning Applications]] > The Task is to devise a precise, step-by-step plan for how we should use our vast data resources to decide which customers should be offered the special retention deal prior to the expiration of their free trial. > Furthermore, this needs to be AB tested to ensure a reduction of monthly churn. ### Problem Definition Currently, monthly churn stands around 60% for the cohort which is too high. ## Constraints This must be done in the context of Diet Doctor monthly subscriptions. That means we will have to use the data available to us in some sort of way. The constraints are: - Limited user information - Age/Gender etc - Analytics - App use / Web use etc - Churned after 30 days - Use transactions data ## Strategy 1. Research -- Medium/Kaggle etc 2. Determine Methodology 3. Prepare a Lab Model 4. Test 5. Insert success into CV ## Research